· Develop and QC TFLs for protocols/reports/manuscripts from RWE research conducted to assess the value of the client's therapies using RWD (e.g. claims and EHR).
· QC programming for descriptive and complex studies using RWD.
· Conduct analyses and develop specifications for descriptive and complex statistics in studies using RWD.
· Write the statistical analysis plan (SAP) for descriptive and complex studies using RWD, including from internal client-sponsored prospective cohort studies, claims, charge master and EHR in collaboration with RWE TA lead.
· Understand methods and programming to support Comparative Effectiveness Research (CER) analyses, as well as analyses of patient-reported outcomes (PRO) or other patient outcome data.
· Work with RWE researchers to generate code lists for new measures in RWD.
Knowledge, Skills and Experience
· Master’s degree (e.g. MA, MSc, MPH) in Biostatistics, Epidemiology or related discipline, such as Outcomes Research from an accredited institution, with a minimum of eight (8) years of relevant, post-graduation experience.
· Doctoral level training with a minimum of two (2) years of relevant experience is preferred. Direct experience in lieu of academic training is acceptable.
· Knowledge of real-world data and experience in observational research study design, execution and communication.
· Strong track record of analysis of a broad range of RWD.
· Formal training in Programming and demonstrated proficiency in statistical analysis programs commonly used in life sciences (e.g. SAS, R).
· Understanding of epidemiology or outcomes research and the application of retrospective or prospective studies to generate value evidence.
· Ability to effectively communicate statistical methodology and analysis results.
· Ability to work effectively in a constantly changing, diverse, and matrix environment.
· Knowledge of US secondary data sources required; additional experience with international data sources is preferred.
· Knowledge and experience in qualitative analysis and data sets (e.g., free-text natural language processing, survey data) is preferred.
